PI3Kγδ inhibitor combined radiation enhances the antitumor immune effect of PD-1 blockade in syngenic triple negative breast cancer models.

e13041 Background: Breast cancer is generally viewed as immunologically cold, imposing an immune-suppressive tumor microenvironment (TME) and responding poorly to lone immune checkpoint blockade (ICB). As an adjunct to ICB, radiation therapy (RT) holds promise in terms of in situ tumor vaccination effect, although it is known to promote immune suppression, increasing regulatory T cells (Treg), myeloid-derived suppressor cells (MDSCs), and M2 tumor-associated macrophages (TAMs). It was our contention that combined use of RT and a PI3Kγδ inhibitor to combat immune suppression might enhance the efficacy of ICB. Methods: Murine breast cancer cells (4T1) were grown in both immune-competent BALB/c mice, humanized patient-derived xenograft (PDX) were established, and both were irradiated by 3 fractions of 24 Gy. A PD-1 blockade (10 mg/kg) and a PI3Kγδ inhibitor (IPI145; 15 mg/kg) were then administered every other day for 2 weeks. Tumors from humanized PDX was sequenced to identify immune-related pathways and to profile infiltrated immune cells. Transcriptomic and clinical data were acquired from The Cancer Genome Atlas (TCGA) pan-cancer cohort, and the deconvolution algorithm was used to profile immune cellular distributions at certain levels (high vs. low) of PIK3Cand PIK3CD expression. Results: In the immune-competent syngenic 4T1 murine tumor model, PD-1 blockade alone led to tumor hyperprogression, whereas a three-pronged strategy of PI3Kγδ inhibitor, RT, and PD-1 blockade significantly delayed primary tumor growth, boosted abscopal effect, and improved animal survival. According to FACS analysis, RT significantly increased not only CD8+cytotoxic T-cell fractions but also immune-suppressive Tregcells, MDSCs, and M2 TAMs. However, PI3Kγδ inhibitor significantly lowered proportions of Treg, MDSCs, and M2 TAMs, achieving dramatic gains in splenic, nodal, and tumor CD8+ T-cell populations after triple combination therapy. There were significantly decreased tumor expressions of p-AKT, PD-L1, and HIF1α by PI3Kγδ inhibition. Triple combination therapy significantly delayed primary tumor growth in humanized PDX model as well and analyses of RNA sequencing data of humanized PDX samples showed decreased immune suppressive pathways with decreased and M2 macrophage and increased CD8+ T-cell by triple combination therapy. In the TCGA pan-cancer cohort, high Treg/CD8+T-cell and M2/M1 TAM ratios and poor overall patient survival was associated with high PIK3CG (PI3Kγ) or PIK3CD (PI3Kδ) gene expression. Conclusions: These findings collectively indicate that PI3Kγ and PI3Kδ are clinically relevant targets in an immunosuppressive TME. Combining PI3Kγδ inhibitor, RT, and PD-1 blockade may thus be a viable approach, helping to overcome the therapeutic resistance of immunologically cold tumors such as breast cancer.